Have Lithium Ion batteries come of age for off-grid systems?

karrak
karrak Solar Expert Posts: 326 ✭✭✭✭
To me, this thread http://forum.solar-electric.com/discussion/351556/batteries-abruptly-stopped-holding-voltage-under-load shows up some of the major problems with lead acid batteries.

  1. It is very difficult to measure how much energy you have stored in a lead acid battery, especially if you draw high power for extended periods due to the poor efficiency of the batteries at high charge and discharge rates (The Peukert effect).
  2. High discharge and charge rates and deeply discharging a Lead Acid battery will reduce its lifespan significantly.
  3. The need for long Absorb times to fully charge the battery
  4. Not keeping a lead acid battery fully charged will decrease its lifespan.

In the past lead acid batteries were really the only economical  alternative so people with off-grid systems just put up with the problems. I don't think this is any longer the case.

I think there is now enough evidence to say that Lithium Iron Phosphate (LFP) and Lithium Manganese Cobalt (LMC) batteries will last at least ten years. With these sort of lifespans and the fact that you can access 80-90% of the energy capacity of the battery on a regular basis makes lithium ion batteries cost competitive with lead acid batteries. There is also now enough equipment available or will be coming available in the next few months that can work with the special needs of Lithium Ion batteries and a number of vendors that will supply lithium ion based systems.

Lithium Ion batteries do not exhibit the problems I have mentioned if you limit charge and average discharge rates to less than half the battery capacity (C/2) and peak discharge rates to less than 2C which is far more power than regular lead acid batteries will supply.

The very high coulomb (current) efficiency of Lithium Ion batteries (>99%), virtually no Peukert effect and low battery leakage means you can accurately measure (to within a few %) the amount of energy stored in an LFP battery by measuring the amount of current going into and out of the battery.

Lithium Ion batteries are quite happy to be charged or discharged at any variable rate up to C/2 (for long life), don't need any Absorb charge time and can be left partially full (partial SOC) with no ill effects.

This makes it far easier and cost effective for people to consider using electricity for climate control, cooking and other power hungry activities.

The one big issue with Lithium Ion batteries is safety and longevity if they are taken out of their safe operating voltage and temperature range. I think this has finally been addressed by the availability of well designed and built battery management systems (BMS).

Simon



Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
modified BMS based on TI bq769x0 cell monitors.
Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
 

Tagged:
«13

Comments

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Does the BMS prohibit charging at low temperatures ?   That's one problem I see with any flavor of Li .

    Another is operator error in setting up charging setpoints and LVD settings,  You mention 2 chemistrys, which will have different voltages and would require someone to manually enter critical information, scaling from 12 to 24 to 48 volts adds to the fun for newcomers,

    Cell Balance  Do cells come in matched sets, or does the user have to test and sort the batteries, tossing out the outliers or limit the pack to the worst cell?  And how to test cells ?    Get this part wrong, just stand back and watch it smoke.

    These are questions neophytes don't know enough about to ask, and will head right down the path to disaster.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    edited January 2017 #3
    @karrak - "Cost effective" to use a battery (any battery) for space heating and cooking. Seriously? Can you run some numbers by us?

    I agree that some batteries are better tban others in certain applications, but I'm having trouble believing any battery is cost effective for heating.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Estragon said:
    @karrak - "Cost effective" to use a battery (any battery) for space heating and cooking. Seriously? Can you run some numbers by us?

    I agree that some batteries are better tban others in certain applications, but I'm having trouble believing any battery is cost effective for heating.
    The two systems I have installed with LFP batteries are both used for cooking. In summer 100% of the total cooking energy and in winter around 20% comes from our offgrid power systems. This is used by microwaves, ovens, induction cooktops, kettles and toasters.

    In winter our total power consumption is around 2.5-3kWh per day which pretty well matches the input from our panels and in summer it is around 5kWh per day. Cooking using electricity makes very efficient use of the extra power we get in summer. We also have not had to spend any money on gas infrastructure or appliances.

    As far as I can see the lifespan of LFP batteries is more dependent on the total charge cycled through them rather than the depth of discharge of each cycle or the average SOC of the battery. This means having a cooking binge which might take our battery down to around 30%SOC once or twice a week will not have a large impact on the overall lifespan of the battery. Our nearly four year old LFP battery shows no sign of any deterioration.

    We use a wood heater for heat and the bulk of the cooking in winter. I wouldn't use an LFP battery for climate control but at a cost of AU$0.23/kWh for the Powerwall 2 which is comparable with our electricity tariff it could make sense for some.

    Simon


    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    mike95490 said:
    Does the BMS prohibit charging at low temperatures ?   That's one problem I see with any flavor of Li .
    I am sure the BMS in the Powerwall or the LG Chem or other lithium ion battery systems by other major players would shut down charging at low temperatures or heat the battery if it is necessary. If you have an electronic/electrical engineering background it is not difficult to implement.
    Another is operator error in setting up charging setpoints and LVD settings,  You mention 2 chemistrys, which will have different voltages and would require someone to manually enter critical information, scaling from 12 to 24 to 48 volts adds to the fun for newcomers,
    Again, this would be done automatically or by the installer in systems like the Powerall 2 etc. For those who have some technical knowledge like allot of people who frequent this forum I don't see it being any more of a problem than setting up lead acid batteries. Most charge controllers these days will automatically detect the battery voltage.

    Cell Balance  Do cells come in matched sets, or does the user have to test and sort the batteries, tossing out the outliers or limit the pack to the worst cell?  And how to test cells ?    Get this part wrong, just stand back and watch it smoke.
    No need to sort the batteries. Buy any batteries from a reputable supplier/manufacturer and they will be matched to within a couple of % or better. Even if there are some duds or the battery pack is not balanced correctly  the BMS will pick these problems up and not let the smoke happen.
    These are questions neophytes don't know enough about to ask, and will head right down the path to disaster.
    I don't make the assumption that all first time posters are morons. For those people with little technical knowledge I would probably suggest that they get a system based around the Powerwall 2, LG chem or another major player installed by a reputable installer. For someone with more technical knowledge I might suggest getting the component parts and installing and commissioning the system themselves.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • mcgivor
    mcgivor Solar Expert Posts: 3,854 ✭✭✭✭✭✭
    edited January 2017 #6
    The future is scary, there will always be those who will revert to old technology to prove that new systems  are not viable, I'm by no means an expert, but don't  the engineers who designed these systems, which currently are not available in North America, but available in other markets , know something about what they are developing and selling outside the United States ? North America is a dangerous market, one mistake could send you to the poor house, which is probably the reason why marketers are reluctant to promote new technologies in such an environment,   My opinions .
    1500W, 6× Schutten 250W Poly panels , Schneider MPPT 60 150 CC, Schneider SW 2524 inverter, 400Ah LFP 24V nominal battery with Battery Bodyguard BMS 
    Second system 1890W  3 × 300W No name brand poly, 3×330 Sunsolar Poly panels, Morningstar TS 60 PWM controller, no name 2000W inverter 400Ah LFP 24V nominal battery with Daly BMS, used for water pumping and day time air conditioning.  
    5Kw Yanmar clone single cylinder air cooled diesel generator for rare emergency charging and welding.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    The one thing that must be done for Offgrid if you want it simple, safe and reliable, is that the Inverter, the Solar controlers, and the BMS need to all communicate.



    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Mangas
    Mangas Solar Expert Posts: 547 ✭✭✭✭
    edited January 2017 #8
    I think OEM manufacturer integrated and networked turnkey off grid systems is the only way to build residential mainstream market acceptability for both small and large applications.

    While there will always be a market for the mix and match engineering types, I'd like to see the day when an installer emails a homeowner's power requirements to the manufacturer's distributor and their entire off grid system is pre-programmed, shipped, and ready to install with off-site program/performance monitoring.

    Getting there.
    Ranch Off Grid System & Custom Home: 2 x pair stacked Schneider XW 5548+ Plus inverters (4), 2 x Schneider MPPT 80-600 Charge Controllers, 2 Xanbus AGS Generator Start and Air Extraction System Controllers, 64 Trojan L16 REB 6v 375 AH Flooded Cel Batteries w/Water Miser Caps, 44 x 185 Sharp Solar Panels, Cummins Onan RS20 KW Propane Water Cooled Genset, ICF Custom House Construction, all appliances, Central A/C, 2 x High Efficiency Variable Speed three ton Central A/C 220v compressors, 2 x Propane furnaces, 2 x Variable Speed Air Handlers, 2 x HD WiFi HVAC Zoned System Controllers
  • softdown
    softdown Solar Expert Posts: 3,812 ✭✭✭✭
    Lots of claims. Lots of promise. No demonstrable, large scale, track record. We shall see.

    I used to subscribe to Popular Science. Must be a "1000 great ideas" for every one that takes hold.
    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries
  • Mangas
    Mangas Solar Expert Posts: 547 ✭✭✭✭
    edited January 2017 #10
    Agree.

    I look at it as "the oxen are slow but the earth is patient", so goes solar development.  Come a long way since I built my system among the naysayers 12 years ago.  Dual zone central A/C. . .?

    But, the grid is still the big dog.
    Ranch Off Grid System & Custom Home: 2 x pair stacked Schneider XW 5548+ Plus inverters (4), 2 x Schneider MPPT 80-600 Charge Controllers, 2 Xanbus AGS Generator Start and Air Extraction System Controllers, 64 Trojan L16 REB 6v 375 AH Flooded Cel Batteries w/Water Miser Caps, 44 x 185 Sharp Solar Panels, Cummins Onan RS20 KW Propane Water Cooled Genset, ICF Custom House Construction, all appliances, Central A/C, 2 x High Efficiency Variable Speed three ton Central A/C 220v compressors, 2 x Propane furnaces, 2 x Variable Speed Air Handlers, 2 x HD WiFi HVAC Zoned System Controllers
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited January 2017 #11
        A little over 8 months ago I installed a whole home hybrid backup system, (doesn't include the dryer or stovetop/oven) consisting of a 4000 watt OB Radian inverter, a classic 150 charge controller, a 3.6kW array and a battery bank of 16 195AH LFP prismatic cells with no BMS installed. This system has met my needs quite well since installation. The rest of this post will be concerning the setup and management of this LFP battery bank.

    Initially I top balanced the bank by fully charging all the cells with a 10 amp 3.2 volt LFP charger. Then connected them all together in parallel for 8 hours. Then installed the cells in the bank, commissioned the system and started daily cycling.
    Yes, I know, so it's more than just a backup system. Would you just let it sit there waiting for a power failure? Not me.
    At first I was checking balance hourly, but the cells were staying in in good balance. Soon it was 2 or 3 times a day, and then daily, all with the same results. The cells continued to stay in balance. Now I check them a couple of times a week.

    After over 8 months time , near the top of charge, say 80%, balance is out 1 millivolt and at 25% SOC balance is out 10 millivolts. I consider this well within my comfort zone and probably won't need to rebalance for some time.
    So, like Simon, and other LFP users (and their are not many) on this and other forums , I agree that when using these cells at low charge and discharge rates of less than C/2, C/3 max for my system, relatively good balance can be maintained.

    Do I recommend professional installations using these cells this way? Absolutely not. Nor do I recommend them to solar newbies. But to someone who is knowledgeable and experienced in this field, I believe it is a viable option for a hybrid or off grid system, especially because of it's ability to handle higher amperage charging, relative to FLA and to rest at PSOC with no ill effects. But, like all the other various battery chemistries, they have their electrical and environmental advantages and disadvantages, and I don't believe one type will answer everyones needs.

    The charging routine is easy. I charge to 55.2 and EA at 10 then float at 54.2 . Absorb takes about 30 minutes. I have the battery efficiency setting in the classic set to 98% to make it work properly with the Wizbang Jr.
    So, I think LA will be around for a while, and many are comfortable with it. All I can say is that my experience with LFP has been good and I want a bigger battery.   :)    

    Rick               
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    edited January 2017 #12
    I have really liked your posts on this Rick ! Really honest about this new battery. 

    I am really happy with the 10KWH LFP I am running for Schneider. It operates safely from 42V to 58.8 V. Really nice graphics in the Bridge hardware/firmware. It does take quite a bit longer than 30 minutes in absorb to fill up 10KWH  :) There is nothing secret about the make I just have not asked them if it can go "out there". It is 1/10 the size and weight of my flooded system with the same capacity if the flooded was used down to 50%, which of course I do. Convection cooling and able to go up to 20KWH anytime.

     I will say if one is concerned about safety, the battery design/manufacture I am testing, is far safer and smaller in my mind than what Tesla is doing with Panasonic. You don't have to write a $500 check to Mr. Musk to get on a list either. Tesla has royally pissed off my supplier also but maybe that is the way they operate.

    The Bridge will be able to match the BMS in different makes of batteries that Schneider feels safe to partner with. Outback is also doing alot in this area for their Skybox.


    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Marc Kurth
    Marc Kurth Solar Expert Posts: 1,142 ✭✭✭✭
    Great synopsis Dave! Thank you.
    I always have more questions than answers. That's the nature of life.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    edited January 2017 #14
    Rick - Raj174,  YES, you *rocked* it!

    From your description, this would be one of the very few cases should you build a larger capacity battery bank, I'd be tempted to take those "used" LiFeP04 cells off your hands.  99% of the time, I'd pass, but given your description - wow.

    I'm also assuming you started with new cells, from a reputable distributor.

    While I don't want to start the bms / no bms war, you demonstrated to a tee what I've been saying about LFP for those with a desire to be at least a little bit involved with their cells.  You fully understood the concept of conservatism, and also the application.  Well done sir.  Just don't get complacent.  Keep your voltmeter handy for regular sanity-checks.

    For lurkers I might add that your single cell LFP charge was a nominal 3.2v charger, that is actually about 3.6 - 3.7v or so absorb.  Just in case someone thinks that charging to 3.2v for initial charge / balancing is sufficient.  I know what you meant, but wanted to throw that in for anyone reading too quickly. :)

    Bravo on your choice of charge voltage, and also termination current.  Not only that, but your choice of float (which is really not absolutely necessary from an on-paper technical standpoint), serves as a backup should a hidden parasitic load occur, and help prevent returning to the garage with zero volt cells.

    I'm assuming of course that you have an LVD in place of some sort.  Hopefully those cells are secured in place, so that an earthquake, or other reason for cell movement won't tweak and torque your connections either sideways, or up and down.  Take a gander at terminal tightness once in awhile, but don't go overboard either and man-handle it.

    Anyway, keep us informed.  So far, that setup should be considered the poster-child on how to do it for the diy'er, that doesn't plan to just abandon them with no regular pm, like simple voltage checks on each cell once in awhile.

    I can't believe it - someone finally did it here diy properly it seems - and at 48v to boot with a simple 16S setup.  Thanks for sharing!

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Forgot - just checking - you have *disabled* any sort of temperature compensation in the Classic 150 right?  With LFP, I surely hope it is disabled!

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Raj174 said:
    After over 8 months time , near the top of charge, say 80%, balance is out 1 millivolt and at 25% SOC balance is out 10 millivolts. I consider this well within my comfort zone and probably won't need to rebalance for some time.
    So, like Simon, and other LFP users (and their are not many) on this and other forums , I agree that when using these cells at low charge and discharge rates of less than C/2, C/3 max for my system, relatively good balance can be maintained.
    One thing to bear in mind with cell imbalance is that it only becomes apparent with LFP batteries within a few % of 100%SOC or 0%SOC unless your battery is very out of balance.

    Below is a graph of the results of some tests I did with a three and a half year old Winston LFP cell to see what the SOC of the cell was when charged at different charge rates. With your charge regime  55.2V (3.45V/cell) and a termination current of 10A? (~0.05C) I would say you are charging your battery to >98%


    When checking battery balance it is important to make the balance check during the absorb part of the charge cycle and preferably at the end of the absorb cycle just before the charge controller switches to float. The graph below shows the voltage difference between the cells in my battery measured today at the end of the charge cycle. The spike near the 10 minute mark is the fridge starting up


    The charging routine is easy. I charge to 55.2 and EA at 10 then float at 54.2 . Absorb takes about 30 minutes. I have the battery efficiency setting in the classic set to 98% to make it work properly with the Wizbang Jr.
    If the efficiency mentioned here is the coulomb (current) efficiency I would set this to at least 99%. My battery's measured coulomb efficiency of ~99.5% agrees with figures in other literature I have read. My understanding is that the efficiency of a new battery will improve with age as a result of the SEI layer stabalising.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭

    It does take quite a bit longer than 30 minutes in absorb to fill up 10KWH
    I generated some graphs of SOC versus absorb time for a three and a half year old Winston cell at different charge rates which I thought maybe of interest.


     I will say if one is concerned about safety, the battery design/manufacture I am testing, is far safer and smaller in my mind than what Tesla is doing with Panasonic. You don't have to write a $500 check to Mr. Musk to get on a list either. Tesla has royally pissed off my supplier also but maybe that is the way they operate.
    I wouldn't have thought that safety would have been an issue. Just look at the number of lithium cobalt (LCO) batteries floating around in laptops, mobile phones etc. etc. etc. which are similar to the lithium manganese cobalt (LMC) batteries used in the Powerwall 2. Having said that I think that LFP batteries are the way to go for offgrid use, especially from a DIY point of view because of the increased safety and also because I think that they should last longer because of the lower cell voltage. The cynical side of me thinks that LMC batteries are perfect for companies wanting you to replace your offgrid or grid connect battery on a more regular basis. I am just hoping that Musk's antics and aggressive pricing will increase the demand for stationary batteries  in general which will enable other suppliers, especially suppliers of LFP batteries to reduce their prices. The last thing we want is a monopoly or small number of battery suppliers left standing dictating how we buy and install their products.

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    I have really liked your posts on this Rick ! Really honest about this new battery. 

    I am really happy with the 10KWH LFP I am running for Schneider. It operates safely from 42V to 58.8 V. Really nice graphics in the Bridge hardware/firmware. It does take quite a bit longer than 30 minutes in absorb to fill up 10KWH  :) There is nothing secret about the make I just have not asked them if it can go "out there". It is 1/10 the size and weight of my flooded system with the same capacity if the flooded was used down to 50%, which of course I do. Convection cooling and able to go up to 20KWH anytime.

     I will say if one is concerned about safety, the battery design/manufacture I am testing, is far safer and smaller in my mind than what Tesla is doing with Panasonic. You don't have to write a $500 check to Mr. Musk to get on a list either. Tesla has royally pissed off my supplier also but maybe that is the way they operate.

    The Bridge will be able to match the BMS in different makes of batteries that Schneider feels safe to partner with. Outback is also doing alot in this area for their Skybox.


    Thanks Dave, I appreciate that. 
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭


    Rick - Raj174,  YES, you *rocked* it!

    From your description, this would be one of the very few cases should you build a larger capacity battery bank, I'd be tempted to take those "used" LiFeP04 cells off your hands.  99% of the time, I'd pass, but given your description - wow.

    I'm also assuming you started with new cells, from a reputable distributor.

    While I don't want to start the bms / no bms war, you demonstrated to a tee what I've been saying about LFP for those with a desire to be at least a little bit involved with their cells.  You fully understood the concept of conservatism, and also the application.  Well done sir.  Just don't get complacent.  Keep your voltmeter handy for regular sanity-checks.

    For lurkers I might add that your single cell LFP charge was a nominal 3.2v charger, that is actually about 3.6 - 3.7v or so absorb.  Just in case someone thinks that charging to 3.2v for initial charge / balancing is sufficient.  I know what you meant, but wanted to throw that in for anyone reading too quickly. :)

    Bravo on your choice of charge voltage, and also termination current.  Not only that, but your choice of float (which is really not absolutely necessary from an on-paper technical standpoint), serves as a backup should a hidden parasitic load occur, and help prevent returning to the garage with zero volt cells.

    I'm assuming of course that you have an LVD in place of some sort.  Hopefully those cells are secured in place, so that an earthquake, or other reason for cell movement won't tweak and torque your connections either sideways, or up and down.  Take a gander at terminal tightness once in awhile, but don't go overboard either and man-handle it.

    Anyway, keep us informed.  So far, that setup should be considered the poster-child on how to do it for the diy'er, that doesn't plan to just abandon them with no regular pm, like simple voltage checks on each cell once in awhile.

    I can't believe it - someone finally did it here diy properly it seems - and at 48v to boot with a simple 16S setup.  Thanks for sharing!

    Thanks PNjunction,
    Yes the cells were new, bought them from an EV car parts company. I recently found a 16S LFP digital cell monitor with alarm, that I will be setting up on the system. I do have HVD and LVD relays set up as redundant failsafe devices using a JLD404AH digital meter. Good point on the charger, it is 3.65.  
    Thanks for all of the posts you have done on LFP cells. They were very informative.
    Almost forgot. Yes, temp comp is disabled.
    Rick
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    karrak said:
    Raj174 said:
    After over 8 months time , near the top of charge, say 80%, balance is out 1 millivolt and at 25% SOC balance is out 10 millivolts. I consider this well within my comfort zone and probably won't need to rebalance for some time.
    So, like Simon, and other LFP users (and their are not many) on this and other forums , I agree that when using these cells at low charge and discharge rates of less than C/2, C/3 max for my system, relatively good balance can be maintained.
    One thing to bear in mind with cell imbalance is that it only becomes apparent with LFP batteries within a few % of 100%SOC or 0%SOC unless your battery is very out of balance.

    Below is a graph of the results of some tests I did with a three and a half year old Winston LFP cell to see what the SOC of the cell was when charged at different charge rates. With your charge regime  55.2V (3.45V/cell) and a termination current of 10A? (~0.05C) I would say you are charging your battery to >98%


    When checking battery balance it is important to make the balance check during the absorb part of the charge cycle and preferably at the end of the absorb cycle just before the charge controller switches to float. The graph below shows the voltage difference between the cells in my battery measured today at the end of the charge cycle. The spike near the 10 minute mark is the fridge starting up


    The charging routine is easy. I charge to 55.2 and EA at 10 then float at 54.2 . Absorb takes about 30 minutes. I have the battery efficiency setting in the classic set to 98% to make it work properly with the Wizbang Jr.
    If the efficiency mentioned here is the coulomb (current) efficiency I would set this to at least 99%. My battery's measured coulomb efficiency of ~99.5% agrees with figures in other literature I have read. My understanding is that the efficiency of a new battery will improve with age as a result of the SEI layer stabalising.

    Simon
    Thanks for the advice Simon, I'll check that out.
    Rick
    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    The efficiency may or may not improve (I don't know yet) but for sure the capacity starts the long slow fade just using the battery.
    Even faster with bad ventilation, temps outside 15 to 30 C, unstable temp/humidity.

    Plenty of danger warnings all over this unit and the gas can kill you in a failure. A type ABC or CO2 fire extinguisher is needed but at this point there is the risk of explosion if the unit is on fire. I am not sure yet what a fireman would do if he knew Li-Ion batteries were hanging on the wall in the garage. Fire Insurance might be an issue. This is all very new in North America. For further info I would look at Australia where a 100 or so of these units have been installed for about 2 years now.

    It is rated IP55 for outdoor installation. :)


    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    edited January 2017 #22
    Rick - sounds good with the JLD and other cell monitoring.  Just remember to keep an eye on things, as even cell monitors can present a parasitic load slightly imbalancing things as time progresses, or worse yet a mosfet failure resulting in a short.    I'm pretty sure you are on top of things though.

    Dave - The gas from an LFP failure would be from a total catastrophe, like 30v per cell burning off the electrolyte - although prior to this would be certain unrecoverable damage / swelling to the cell to begin with.    Compare this to about 4.5v for a non-lfp cell. :)  Big difference.  Looks like Rick has taken measures to do things properly.  But you bring up a good point:

    That's why I wonder why a slightly larger, but inherently safer powerwall isn't made from LFP, instead of the other chemistries.   I know I'd rather have LFP in the garage, and maybe insurance companies would too - provided they knew the difference between chemistries in the first place.  Oh well, that's a whole different consumer market than what we do here.

    We return once again to the motorcycle LFP batteries sitting a few inches away from your rump, (and also in high-performance wheelchairs).  Any other lithium chemistry that close to your cheeks is a no-go. :)  There's venting (LFP), and "venting with flame" (non-lfp) abuse.  Of course one's own wiring infrastructure, if unsafe and foolishly engineered, can catch fire.

    One can also blow the tops off their flooded batteries spewing acid and hydrogen gas too - so like with ALL batteries, safety should be a foremost consideration, be it fire-extinguishers or whatever else is common-sense when working with any battery bank.

    SAFETY - two somewhat recent memorable experiences with my lfp bank:

    1) Cleaned the terminals and links from oxidation with a non-soapy and non-metallic pad.  Applied *light* coating of anti-oxidant paste.  (We are dealing with a lot of dissimilar metals here like copper, aluminum, nickel etc), and proceeded to blow out any dust from the sunken terminal threads with a camera blower bulb.  RIGHT INTO MY EYES.

    2) Wrapped all tool handles used to construct the bank with electrical tape right up to the heads of the tools.  So far so good.  But I was wearing metal-frame reading glasses and it was summer.  Glasses shot off my head when looking down and bounced off terminals.  Fortunately, it did so without simultaneous contact with the terminals.

    That's when I figured it was time to SLOW DOWN, and get some goggles on.  Stuff happens when you least expect it.
     


  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭

    Raj174 said:
    I recently found a 16S LFP digital cell monitor with alarm, that I will be setting up on the system.
    Rick, could you give some details on the 16S digital cell monitor you have found. I am interested in finding something better than the venerable but flawed Cellog 8 to recommend to others.

    Thanks
    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    The efficiency may or may not improve (I don't know yet) but for sure the capacity starts the long slow fade just using the battery.
    Even faster with bad ventilation, temps outside 15 to 30 C, unstable temp/humidity.

    Plenty of danger warnings all over this unit and the gas can kill you in a failure. A type ABC or CO2 fire extinguisher is needed but at this point there is the risk of explosion if the unit is on fire. I am not sure yet what a fireman would do if he knew Li-Ion batteries were hanging on the wall in the garage. Fire Insurance might be an issue. This is all very new in North America. For further info I would look at Australia where a 100 or so of these units have been installed for about 2 years now.

    It is rated IP55 for outdoor installation. :)



    This video https://www.youtube.com/watch?v=pxP0Cu00sZs, which is one of PNjunction's favorites gives quite allot of information on lithium ion battery coulomb efficiency and other interesting stuff. There are some graphs of efficiency for the different lithium ion chemistries around the 27 minute mark.

    I hope I don't come across as being flippant when it comes to lithium ion battery safety. Maybe I am a little tired of the naysayers implying that lithium ion batteries will explode if you just look at them. My understanding is that if they are kept within their safe operating voltage, current and temperature range that they are safe. It is when you take them outside the safe operating zone and then continue to use them that you can run into problems. This is why I am so keen on having some sort of BMS to make sure that the battery stays within its safe operating range.

    Simon

    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Interesting thread. I'm considering LFP for an upcoming replacement of house bank on my boat and eventually (hopefully some years off) for my cabin banks. Low temps might be a problem though. I know low temp (<0°C) charge/discharge is bad, but is cold storage ok?

    My guess on the insurance thing is there will be a few high profile fails and insurance cos may increase premiums substantially for premises with any type of large lithium bank until they have solid empirical evidence that any are safe. They may even refuse individual cover completely forcing vendors to include coverage in warranty.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    I thought I would post some more graphs of battery balance, this time showing how having a long absorb time can effect how large a voltage difference you can get.

    This graph is from data I downloaded from my friend's system today. Zero minutes is the start of the absorb phase of the charging. His solar controller only allows one to program the absorb time with a minimum of one hour. It puzzles me why it continued the absorb phase for around two hours when it is programmed for one. Maybe it is something to do with the fact that the battery is in use during the absorb phase.

     The second graph shows how the individual cell voltages diverge, some actually go down!

    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Dave Angelini
    Dave Angelini Solar Expert Posts: 6,728 ✭✭✭✭✭✭
    I really have very little opinion yet as I am just reading the 9 different warning stickers on the unit. Earthquake and vibration or denting the case are really big stickers. Also as with the Tesla, you probably would not see any physical sign of a bad battery as the case of the unit would hide it. There are more things to go wrong and so that is why my opinion for offgrid is still being formed. The size is amazing but after the wow wears off one realizes that for most offgrid homes, size is not a big deal. The warranty for cycles is great but the percentage of what you pay for the unit is down to 30% by year 5. The efficiency is great but again  a few extra solar panels and that issue is moot. It really comes down to maintenance vs cost vs reliability so far.

    I am glad there is a BMS built into the unit with a USB drive for firmware update and that the Bridge is also checking on the BMS with it's own firmware.
    Without a BMS,  I remember a client in Baja who got hit with lightning after a hurricane. His charge controller shorted the PV in to output and for 3 days the battery had close to full PV in on it. It just boiled off the water in that case.

    "we go where power lines don't" Sierra Nevada mountain area
       htps://offgridsolar1.com/
    E-mail offgridsolar@sti.net

  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited January 2017 #28
    karrak said:

    Raj174 said:
    I recently found a 16S LFP digital cell monitor with alarm, that I will be setting up on the system.
    Rick, could you give some details on the 16S digital cell monitor you have found. I am interested in finding something better than the venerable but flawed Cellog 8 to recommend to others.

    Thanks
    Simon
    Sure Simon, here's a link to where I bought it. I think they are based in Hong Kong.

    http://www.dhgate.com/product/high-accuracy-battery-management-system-bms/375809910.html

    This is the manual.



    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • Raj174
    Raj174 Solar Expert Posts: 795 ✭✭✭✭
    edited January 2017 #29
    Estragon said:
    Interesting thread. I'm considering LFP for an upcoming replacement of house bank on my boat and eventually (hopefully some years off) for my cabin banks. Low temps might be a problem though. I know low temp (<0°C) charge/discharge is bad, but is cold storage ok?

    My guess on the insurance thing is there will be a few high profile fails and insurance cos may increase premiums substantially for premises with any type of large lithium bank until they have solid empirical evidence that any are safe. They may even refuse individual cover completely forcing vendors to include coverage in warranty.
    Hi, Estragon,
    From what I have learned, cold storage is fine. The colder the better, down to -20 degrees C, as it slows down self discharge.
    I found a boating forum that has a lithium battery section that may provide more information of interest.

    http://www.cruisersforum.com/forums/f14/is-lithium-iron-phosphate-the-future-of-marine-batteries-157428.html

    Rick

    4480W PV, MNE175DR-TR, MN Classic 150, Outback Radian GS4048A, Mate3, 51.2V 360AH nominal LiFePO4, Kohler Pro 5.2E genset.
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Estragon - since your application also contains a marine diy aspect, the total forum whopper thread about using LFP for housebanks starting from 2011 (still active!) can be found at the same forum:

    http://www.cruisersforum.com/forums/f14/lifepo4-batteries-discussion-thread-for-those-using-them-as-house-banks-65069.html

    I'd also heartily recommend another marine-based LFP support page by another heavy-hitter in the field:

    https://secure2.pbase.com/mainecruising/lifepo4_on_boats&page=1

    The above can also guide you in regards to specific marine safety specifications / alternator wiring setups, etc on the other project docs.

    These two resources from what has been discussed here are probably the closest 3rd-party info we can get to using LFP in our solar application.  EV forums are also helpful, and there are quite a few I respect there as well, however their application (max capacity / max range) have to be taken into account.

    As always, be on the lookout for ops using used and abused cells from the start of any thread.  Most of those who get burned, despite following proper operating procedures, usually fess up to going the cheap route and paying the piper for it.


  • Estragon
    Estragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Thanks. A lot to learn on these yet.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter